584
Growth and Eicosapentaenoic Acid Production by Phaeodactylum
tricornutum in Batch and Continuous Culture Systems
Wichien Yongmanitchai and Owen P. Ward*
Department of Biology, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1
Maximum specific growth rate {t~m~)of Phaeodactylum
tricornutum increased with increasing culture reactor
surface-to-volume ratio. Values for ~ of 0.647, 0.377
and 0.339 day -x were observed for the 7~mL tube, 5.~L
tank and the 1C~Ltank, respectively. Higher biomass was
achieved in the 75-mL batch culture tube under con-
tinuous light as compared with light cycle conditions.
Palmitic acid, palmitoleic acid and eicosapentaenoic acid
(EPA) accounted for over 60% of total fatty acids in the
batch tube culture, with EPA content increasing to a max-
imum after three days. In chemostat cultures, run at dilu-
tion rates of 0.15 day -1 (0.45 of/~mJ and 0.3 day -1 (0.9 of
~ ) , cell concentration reached a steady state of 2.18
and 0.7 g/L, respectively, while contents of EPA per liter
of culture at steady state were 100.9 and 82.5 mg/L,
respectively. At both dilution rates, EPA content of total
fatty acids was the same (35.0-35.2%). At a dilution rate
of 0.3 day -i , the continuous culture system manifested
productivities of 0.51 g/L/d and 25.1 mglL/d for biomass
and EPA, respectively.
KEY WORDS: Algae, eicosapentaenoic acid, omega-3, Phaeodac.
tylum tricornutum, polyunsaturated fatty acids.
Omega-3 fatty acids, eicosapentaenoic acid (EPA) and
docosahexaenoic acid (DHA), have been shown to have
substantial beneficial effects on human health (1). Evi-
dence suggests that these acids have potential for use in
prevention or treatment of heart and circulatory diseases,
inflammatory problems and cancer (2). The positive effects
of omega-3 fatty acids in human physiology were first
observed in populations dependent primarily on fish diets
or human subjects fed on diets of fish oils. These polyun-
saturated fatty acids are primarily produced by marine
microorganisms upon which the fish feed (3). There is
substantial interest in seeking alternative production
sources of EPA and DHA from algae and fungi because
of concerns regarding fish oil supply and the complex
problems of purifying these omega-3 fatty acids from
crude fish oils (4).
In an earlier investigation, we described the effects of
nitrogen source, phosphate, sodium chloride, growth fac-
tors, precursors, CO2, temperature, initial pH and inocu-
lum size on biomass and EPA production by a freshwater
algal strain of Phaeodactylum tricornutum in batch cul-
tures (5). In this paper, the medium composition and op-
timum culture conditions observed have been used to in-
vestigate kinetics of growth of the diatom P. tricornutum
in batch and continuous cultures.
MATERIALS AND METHODS
Organism. Phaeodactylum tricornutum UTEX 640 was
obtained from the culture collection of Algae at the
*To whom correspondence should be addressed.
JAOCS, Vol. 69, no. 6 (June 1992)
University of Texas at Austin. The culture was maintained
and propagated as previously described (5).
Culture conditions. For both batch and continuous
culture studies, composition of the medium and other con-
ditions were based on earlier culture optimization studies
(5) and consisted of 5.0 g/L NaC1, 0.7 g/L urea, 1.2 g/L
MgSO4"7H=O, 0.6 g/L KC1, 0.3 g/L CaC12, 0.1 g/L
K2HPO4, 0.55 g]L NaHCOs, 30 mg/L Na 2 EDTA, 6 mg/L
HsBO 3, 2 mg/L FeSO4"7H20, 1.4 mg/L MnC12, 3.3 mg/L
ZnSO4"7H20, 7.0 #g/L Co(NOs)2"6H20 and 2.0 flg/L
CuSO4"5H20. Initial pH was 7.6 and culture temperature
was 20 +__IC.Three types of culture units were used: the
original 75-mL culture tube, a 5.6-L glass tank (dimen-
sions 30 18 X 12 cm) and a 16-L glass tank (dimensions
40 25 20 cm). For tank cultures, temperature control
was provided by a cooling coil and a heater with ther-
mostat. Two submersible circulating pumps were used to
ensure uniform temperature distribution and homogeneity
of the culture. Two sets of double fluorescent lamps
(GRO-LUX, Sylvania, F40712-GRO-WS) were provided
laterally at both sides of the tanks to supply light inten-
sity of ca. 4000 lux. Unless otherwise stated, a photo-
period of 16 h light to 8 h dark was used throughout the
experiment. Air, supplemented with 1% carbon dioxide,
was filtered through a Microfibre disposable filter unit
(Grade AQ, Balston, MA) and then supplied to the culture
at the outlet of the circulating pump which generated fine
air bubbles to enhance gas exchange. Aeration rate was
1 vol of air per volume of culture per minute (VVM). For
continuous cultures, a peristaltic pump with calibrated
flow rate was employed to feed fresh nutrient medium into
the culture system. Another pump attached to the level
control apparatus was operated at higher flow rate than
the feed pump to withdraw culture broth and to maintain
constant volume
Determination of growth parameters. For dry weight
determinations, samples were taken at the middle of the
light period. Culture broths (50 mL) were filtered through
0.8-#m membrane filters and washed twice with 50 mL of
saline solution. Cells were dried at 60C to constant
weight. Specific growth rate and doubling time were
calculated from the equations # = 2.3(1og Xt - log X0)/t;
T = 0.693/~ (6), where/~ = specific growth rate, d-l; X =
cell concentration at time t, cells/mL; X 0 = cell concen-
tration at time 0, cells/mL; t = time, d, and T = doubling
time, d.
In chemostat continuous cultivation, at steady state
there is no change in cell concentration over time and
sterile feed, so that ~ = D (dilution rate, day -I) and
D = F/V (F is medium flow rate, mL/day; V is culture
volume, mL).
Lipid analysis. Methods for extraction and analysis of
lipids have been described previously (5).
RESULTS
The patterns of biomass production in batch cultures with
the different size culture units are illustrated in Figure 1.
 585

GROWTH AND EPA PRODUCTION BY P. TRICORNUTUM

3
2
J ~. , , ~ . o . 0 O . ~ -0" . . . .

0.5

~9 0.3
" 16LT.
0.2
I I ,, I I I I I l I I I
0.1

2x108
108

v
5x107 J . .~. & ~ A ~
f . . ~.- ~ . . . . . o- .... o. . . . . o
2x107 J ~ ~ _ . . 0 . o . . 0 "-
Z 107 -"

rD
5xlO6

_ I ! 1 1 , ! 1 ! f ,, I l I
2x106
0 i 2 3 4 5 6 7 s 9 i0 n 12
Time (days)
FIG. 1. Pattern of biomass production in batch cultures with different size culture units.

The smaller the culture unit, the higher the growth rate
observed, and biomass dry weight values achieved after
11 d were 6.12 g/L, 2.67 g/L and 1.65 g/L for the 75-mL
tube~ the 5.6-L tank, and 16-L tank, respectively. In our
study, the culture tube had the highest surface-to-volume
ratio of approximately 1 cm -1, wl~le the corresponding
ratios for the small and large t~m'-s were 0.193 and
0.125 cm -~, respectively. As far as growth behavior is
concerned, there was no apparent lag phase in any of the
three culture systems. In the 75-mL tube the culture ex-
hibited a maximum specific growth rate (~max) of
0.647 day -1 (doubling time of 1.071 day). For the 5.6-L
tank, ~ of 0.377 day -~ (equivalent to a doubling time
of 1.839 days) was observed on the third day of incuba-
tion. For the 16-L tank, ~ was 0.339 day -1 (doubling
time, 2.044 days) occurring on day 1. The changes in
growth parameters, biomass, ~ and t, of P. t r i c o r n u t u m
in the three culture units are presented in Figure 2. A semi-
logarithm plot of biomass vs. time revealed that P. tricor-
n u t u m followed an exponential pattern of growth when
light was saturated in the culture tube. Linear growth ap-
peared to be prominent, particularly at high cell density
at the later stages of the growth cycle.
When the organism was cultured in the 75-mL t u b s
under continuous light, a significant improvement in
growth was observed, with cell biomass reaching 7.31 g/L
compared to 5.26 g/L for light cycle conditions after 10 d
(Fig. 3).
When the fatty acid composition of P. t r i c o r n u t u m was
monitored throughout the growth cycle in 75-mL tube
cultures, palmitic acid (16:0), palmitoleic acid (16:1) and
EPA were the three predominant fatty acids formed,
accounting for over 60% of total fatty acids (Table 1).
Amounts of C16 acids were higher in the early growth
phase and decreased gradually in the later stages of
growth. In contrast, EPA content was lower during the
lag phase and increased to a maximum value after three
days. After four days of cultivation, proportions of the dif-
ferent fatty acids remained relatively constant (Fig. 4).
To study the kinetics of growth of R tricornuturn in con-
tinuous culture, the 5.6-L tank was used as a model system
due to constraints in controlling feed rate and limitations
on culture sampling for analysis encountered in the tube
culture unit. Chemostat culture was the choice of opera-
tion because of its simplicity. Two levels of dilution rate
(D), i.e. 0.15 day -1 (840 mL/d) and 0.3 day -I (1680 mL/d),
corresponding to 0.45 and 0.9 of ~ , respectively, were
chosen to avoid the possibility of wash-out at higher D.
The time courses of biomass and EPA production in these
continuous cultures are presented in Figures 5a and 5b.
As the desired product, EPA, was directly related to
biomass [Type 1 fermentation according to Luedeking (7)],

JAOCS, Vol. 69, no. 6 (June 1992)

586

W. Y O N G M A N I T C H A I A N D O.P. W A R D

10
a. 7 5 - m l - T u b e ~ .... o
*
0.8

A
0.6

0.4
.- 0

0.2

f I I I I f I I l 1 0
0
,-J b. 5 . 6 - L - T a n k Daw .2
v 10 1
e.) Sp. G ro.~..hRate
**
~ Doub.li~g..~me ..
E-, 8 0.8
e,o 0 ca

,
6 0.6
o
C~
4 ,,- A ~ ~ A~ " . .~"
0.4 t.j

.... ... .* "-3

v 2 -O "O . . . . "O .... 0.2
3:
C3 0 0
c. 1 6 - L - T a n k
0.8

P
0.6

.~. . 0 . . .*~
0"*" 0.4

0.2

0 0
0 1 2 3 4 5 6 7 8 9 10 I1 12

Time (days)
F I G . 2. C h a n g e s in growth parameters as a function of t i m e in the three culture units.

JAOCS, Vol. 69, no. 6 (June 1992)

 587

GROWTH AND EPA PRODUCTION BY P. TRICORNUTUM

5
3
2

0.5
0.3
0.2

I I I , I I I ! ! I ! I
0.1

2xlO 8

- 108 / ..~ s " ~ ""

5xlO 7

/ /At ContinuousLight
--X 2xlO 7 / // O
"6
L9 J .. ~,' L/ght Cycle
10 7 _/_ A " "'&-"

5xlO6 t t I ................. I I I I I I I I
0 1 2 3 4 5 6 7 8 9 10 ll 12

Time (days)
FIG. 3. Production of biomass with time under continuous light and light cycle conditions in 75-mL tube
batch cultures.

TABLE 1 0.03 g/L after the second day of fresh medium introduc-
tion. E P A content (% of total f a t t y acids} and E P A pro-
Changes of Fatty Acid Composition in P. tricornutum UTEX 640
During a Growth Cycle in 75-mL Tube Cultures duction (mg/L} were also maintained at constant levels of
35.0 +_ 0.70% and 100.9 +_ 2.48 mg/L, respectively. The
Fatty Fatty acid content, % of total fatty acids same phenomenon was observed at the higher dilution
acid Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 rate (D----0.3 day -1) but with lower values of biomass
concentration (1.7 __ 0.04 g/L), and E P A content of the
16:0 21.0 15.8 13.3 10.4 10.1 10.1 culture {82.5 + 2.35 mg/L} with a similar proportion of
16:1 32.1 26.7 23.1 19.7 20.4 21.1 E P A in total f a t t y acids {35.2 _ 0.73%}. A comparison of
18:0 0.7 0.5 0.3 0.4 0.3 0.3 the production efficiency of the continuous culture
18:1 2.7 2.0 2.4 1.7 1.7 1.3 systems indicated that, at a dilution rate of 0.3 d a y -1,
18:2 1.9 1.5 2.4 3.3 3.2 3.7
18:3 0.9 0.6 0.9 0.8 1.0 1.0 productivities for biomass and E P A were 0.51 g/L/d and
20:4 0.6 0.3 0.4 0.6 0.9 1.1 25.1 mg/L/d, respectively. Corresponding values for bio-
20:5 19.0 24.2 29.9 28.7 31.4 32.5 mass and E P A at a dilution rate of 0.15 day -1 were
22:6 2.5 5.3 8.9 6.2 4.5 3.7 0.327 g/L/d and 15.14 mg/L/d, respectively.
Others 18.7 23.0 18.4 28.2 26.7 25.2
DISCUSSION

The observation t h a t continuous light exposure resulted

fresh medium was fed to the culture tank at the t e n t h day in higher biomass production t h a n light cycle conditions
where ~ had been passed to ensure m a x i m u m initial is consistent with cultivation studies with Porphyridium
cell concentration. Furthermore, at this period E P A ac- strains. H i g h growth rates were observed when Porphyri-
cumulation in the cells had already attained highest value dium cultures were grown under continuous light, and no
{approximately 35% w/w of total f a t t y acids}. At D -- 0.15 distinct requirements for a specific light-dark regime were
day -1, cell concentration reached a steady state of 2.18 _ observed {8,9}. In contrast, Brand and Guillard riO} have

JAOCS, Vol. 69, no. 6 (June 1992)

588

W. YONGMANIqX3HAI AND O.P. WARD

40

35

o 30 ~t t ",.,,
.~ ,#~,, _ .....o-- - -o

"~ 20
.9,o
]5 o..

IO 0 ..... O ..... ~" .... ~ ..... 0 ..... 0* .... "13 . . . . . 0

5 I I 1.... I I | I f I I I

/ %
50 t
!
\

0 ~ .It\ x

o~ : .. ~ "13-. ~ _ B _ _ 6 ~ -0" -_B ~-

: t~.

~ 20
Lr~ / "0..
""0 ..... 0 ..... 0 ..... O ..... 0 ..... ..... 0 ..... 0
10

0 I I I I I I I I I ........ I I

200

150

I00 .- "" "

0
0 I 2 3 4 s 6 7 s 9 ]o n 12
Time (days)
FIG. 4. Patterns of production of palmitic (16:0), palmitoleic (16:1) and eicosapentaenoic (20:5) acids by P.
tricornutum in 75-mL tube batch cultures.

JAOCS, V o l . 69, no. 6 ( J u n e 1992)

 589
GROWTH AND EPA PRODUCTION BY R TRICORNUTUM

120
(a) D = 0.15 da] 1
100

f-n

8o
g v

3O _ Z~" A'" "" 3 o

v 20
i" 4o
"y~'" / ,El ,El_ B _El. B ..1~. B .~3_1=1_{3_B _!:3, FI ..,El_B _El
2 =0
~J
o
10 20
<

I I I I ..... I ..... I 1 I I I I ........... | I

0 O

(b) =0.30Dd a y - l ~
100

Q. so 4 a~

~3..O. y O O.u - - ~ v ~0
30 ~ 60 3 "3
O'~" /
o ~

v 20 4o 2 --;
U
A ~ Feed start ~- -
o
10 20 El,"1~
<
~J ~cr a~
0 , I,, I I I I I I I, I , I I I ,,I ...... I I
0
0 2 4 6 8 10 12 14 16 I8 20 Z2 24 25 28 30 32
Time (days)
FIG. 5. Kinetics of growth and E P A production by P. tricornutum in chemostats at dilution rates of (a) D = 0.15 day -1 and {b) D =
0.30 day -1.

shown that many micro-algae will not grow in continuous (6 cm diameter) polyethylene tubes in which the culture
light and need a dark period. In most laboratories, light broth was circulated, has also been used in pilot-scale pro-
regimes of 12-14 h and dark periods of 10-12 h are typi- duction of Porphyridium cruentum (9).
cally used for algal cultivation. The finding that increas- In growth studies with marine species of R tricornuturn,
ing algal growth rates were observed with increasing reac- Siron et al. (14) noted that C16:0 and C16:1 fatty acids
tor surface-to-volume ratios has important implications increased toward the end of the growth period with a loss
for reactor design. Vertical glass tubular reactors were of EPA, while Arao et al. (15) reported the opposit~ Our
used to grow several species of algae under laboratory con- observation that EPA content increases at the later stages
ditions by Miyamoto et aL (11), providing high surface- of growth at the expense of C16 fatty acids is therefore
to-volume ratios similar to our tube culture Outdoor mass consistent with the findings of Arao et aL {15). However,
cultivation of micro-algae was usually conducted in open in two green algae, Chlorella vulgaris and Scenedesrnus
raceway ponds where maximum light intensity may be ob- obliquus, larger amounts of polyunsaturated C16 and C18
tained by limiting the culture depth (12) This type of re- fatty acids were observed during the initial growth stages
actor is being used successfully for the cultivation of while mainly saturated fatty acids were produced at the
Chlorella, Spirulina, Porphyridium and other algae in the end of the growth stage (16).
United States, Israel and Thailand (13). Another type of Growth rate of our freshwater P. tricornutum species at
reactor, consisting of an assembly of long (20 m), thin 0.647 day -1 was comparable to that of marine species

JAOCS, Vol. 69, no. 6 (June 1992)

590
w. YONGMANITCHAI AND O.R WARD
reported at 0.14 to 0.87 day -1 in nutrient-deficient
medium (17). When compared to P. cruentum (18), whose
m a x i m u m specific growth rate reached 1.39 day -1, our
culture seemed to be inferior, but E P A content of P. trico~
n u t u m was higher (3.3% w/w of dry weight compared to
2.1% w/w).
Chlorella m i n u t i s s i m a had E P A productivity of 3.01
mg/L/d (19) compared to 19.0 mg/L/d for our culture (5).
The filamentous fungus Mortierella alpina produced
0.3 g/L of EPA (27 mg/g dry mycelia) equivalent to approx-
imately 50 mg/L/d (20).
When compared with other algae of economic potential,
R tricornuturn performed relatively well in terms of
b i o m a s s production. For example, with Chlorella
pyrenoidosa and C ellipsoidea growth rate ranged from
0.16 to 0.50 day -~ depending upon carbon and nitrogen
source (21). Spirulina platensis manifested a specific
growth rate ranging from 0.11 to 0.30 day -~ depending
on culture temperature (22). Scenedesrnus sp., a unicellular
green alga, had a specific growth rate as high as 0.65
d a y -~ under favorable g r o w t h conditions (13). The
relative growth c o n s t a n t of P. cruenturn, a unicellular red
alga, was found to v a r y between 0.56 and 1.17 day -1
depending upon culture conditions, particularly light in-
tensity and CO2 supply (23}.
ACKNOWLEDGMENTS
Support for this research by the Natural Sciences and Engineering
Research Council of Canada is gratefully acknowledged. O. R Ward
is holder of an NSERC Industrial Research Chair, co-sponsored by
Allelix Biopharmaceuticals Inc, Canad& Support for W. Yongmanit-
chai from the Canadian International Development Agency is also
gratefully acknowledged.
REFERENCES
1. Dyerberg,J., H.O. Bang, E. Stoffersen, S. Moncada and J.R. Vans
Lancet ii:117 (1978).
2. Simopoulos, A.P., J. Nutr. 11~.521 (1989).
JAOCS, Vol. 69, no. 6 (June 1992)
3. Ackman, R.G., in Nutritional Evaluation of Long-Chain Fatty
Acids in Fish Oi~ edited by S.M. Barlow, and M.E. Stanby,
Academic Press, London, England, 1982, pp. 25-28.
4. Yongmanitchai,W., and O.R Ward, Proa Biochem. 24:117 (1989}.
5. Yongmanitchai, W., and O.R Ward, Appl. Environ. Microbiol.
57".419 {1991}.
6. Vonshak, A., in CRC Handbook of Microalgal Mass Culture
edited by A. Richmond, CRC Press, Inc, Boca Raton, FL, 1986,
pp. 117-145.
7. Luedeking, R., in Biochemical and Biological Engineering
Scienc~ edited by N. Blakebrough, Academic Press, London,
England, 1967, pp. 181-243.
8. Vonshak, A., Z. Cohen and A. Richmond, Biomass 8:13 {1985}.
9. Vonshak, A., in Micro-algal Biotechnology, edited by M.A.
Borowitzka, and L.J. Borowitzkm Cambridge University Press,
Cambridge, England, 1988, pp. 122-134.
10. Brand, L.E., and R.R.L. Guillard, J. Exp. Mar. Biol. EcoL 5~119
(1981).
11. Miyamot~ K., O. Wable and J.R. Benemann, BiotechnoL Lett.
1~.703 (1988).
12. Dodd, J.C., in CRCHandbook of MicroalgalMass Culture edited
by A. Richmond, CRC Press, Inc, Boca Raton, FL, 1986, pp.
265-283.
13. Richmond, A., IbicL, 1986, pp. 199-243.
14. Siron" R., G. Giusti and B. Berland, Mar. Ecol. Prog. Ser. 55:95
(1989).
15. Arac~ T., A. Kawaguchi and M. Yamada, Phytochem. 26.'2573
(1987).
16. Piorreck, M., and P. Pohl, Ibid. 2&217 (1984).
17. Osborne,B.A., and R.J. Geider, Plant, Celland Environment ~.617
(1986).
18. Cohen, Z., in Proceedings World Conferences on Biotechnology
for the Fats and Oils Industry, edited by T.H. Applewhite,
American Oil Chemists' Society, 1988, pp. 285-287.
19. Seto, A., H.L. Wong and C.W.Hesseltine, J. Am. Oil Chem. Soa
61:892 (1984).
20. Shimizu, S., Y. Shinmen, H. Kawashima, K. Akimoto and H.
Yamada, Blochem. Biophys. Res. Comm. 150.'335 (1988).
21. Samejima, H., and J. Meyers, J. Gen. MicrobioL 18:107 (1958).
22. Richmond, A., A. Vonshak and S.M. Arad, in Algal Biomass,
edited by G. Shelef, and CJ. Soeder, Elsevier]North-Holland
Biomedical Press, Amsterdam, Netherlands, 1980, pp. 65-72.
23. Jones, It.E, H.L Speer and W. Kury,PhysioL Plana 16:636(1963).
[Received November 19, 1991; accepted March 18, 1992]